Developer tools

AI agent frameworks: the tech stack of 81 real products

Libraries and SDKs for building agents and LLM apps — orchestration, tools, multi-agent.

What sets them apart

Picks at least twice as common here as among products overall, in decisions at least 10 of them show.

The typical stack

The leading pick where at least 8 of them show the decision and the leader has at least a quarter of it.

DecisionMost common pickShareAt default usage
LLM APIOpenAI API64 of 69 · 93%$10/mo · GPT-6 Luna
DatabasePostgreSQL32 of 50 · 64%—
Backend FrameworkFastAPI38 of 49 · 78%—
Frontend FrameworkReact37 of 49 · 76%—
AI SDK & Agent FrameworkLangChain21 of 39 · 54%—
Vector DatabaseQdrant20 of 33 · 61%$103/mo · Standard (3 nodes, 0.5 vCPU / 4 GiB each)
HostingGitHub Pages10 of 29 · 34%—
Database Access & ORMsSQLAlchemy20 of 26 · 77%—
Transactional EmailResend14 of 19 · 74%$20/mo · Pro 50k
LLM Observability & EvalsLangfuse7 of 17 · 41%$29/mo · Core
File StorageAmazon S37 of 14 · 50%$11/mo · S3 Standard (US East)
Product AnalyticsPostHog14 of 14 · 100%$0/mo · Pay-as-you-go
Web AnalyticsPlausible Analytics5 of 12 · 42%$19/mo · Starter 100k
AuthenticationClerk4 of 10 · 40%$0/mo · Hobby
Background Jobs & CronTemporal3 of 8 · 38%$5.50/mo · Developer (pay as you go)

Monthly bills add up to about $197 at the calculators' default usage, list prices. Set your own usage →

What they chose, decision by decision

Among the AI agent frameworks that show each choice, from makers' products and open-source code alike.

Small samples, fewer than 8 products: Error Monitoring (Sentry 6) · Image & Video Hosting (sharp 6) · Payments (Stripe 5, Autumn 1) · Lifecycle Email (Loops 5)

AI agent frameworks we track

38 makers' products and 43 open-source projects. Makers' products first.

NotteBuild and deploy reliable browser agents at scale+3
ThesysC1 is an API that makes LLM responses come back as interactive UI components such as charts, forms and cards instead of plain text. For developers building AI applications.+1
VoltAgentBuild TS AI agents with n8n-style observability+1
Lamatic.aiLamatic is a platform for building AI agents in a visual editor and deploying them as serverless APIs, with a built-in vector database, tracing and evaluations. It is for development teams shipping agent features.
Superexpert.AIOpen-source platform for enterprise AI agents, web-first.
GraphBitGraphBit is an open-source framework for building AI agents, with a Rust core and a Python API. It is for developers building multi-agent workflows.
WordwareWeb-based IDE for building LLM applications, where prompts written in natural language are combined with loops, logic and function calls. For cross-functional teams building AI apps.
FinyuusA code-first language for durable, governed AI workflows
Conva.AIBuild an AI Assistant for your App in 1 click
LyzrPrivate Agent SDKs to build Generative AI Applications
CrowMake your product AI-native via AI Copilot
tamboGenerative UI SDK for React+18
agentfieldBuild, run and scale AI agents like API and microservices+15
hatchet🪓 An orchestration engine for background tasks, AI agents, and durable workflows+7
ag2AG2 (formerly AutoGen): The Open-Source AgentOS.Join us at: https://discord.gg/sNGSwQME3x+4
workflowWorkflow SDK: Build durable, reliable, and observable apps and AI Agents in TypeScript+16
ag-uiAG-UI: the Agent-User Interaction Protocol. Bring Agents into Frontend Applications.+12
eveThe Open Framework for Building Agents+8
omnigentOmnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and c+8
pyspurA visual playground for agentic workflows: Iterate over your agents 10x faster+8
ruflo🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, feder+8
dynamiqDynamiq is an orchestration framework for agentic AI and LLM applications+7
eidolonThe first AI Agent Server, Eidolon is a pluggable Agent SDK and enterprise ready, deployment server for Agentic applications+7
Vision-AgentsOpen Vision Agents by Stream. Build voice and vision agents quickly with any model or video provider. Uses Stream's edge network for ultra-low latency.+4
griptapeModular Python framework for AI agents and workflows with chain-of-thought reasoning, tools, and memory. +3
lightspeedDeterministic agent harness for Temporal (in Rust)+3
agent-squadFlexible and powerful framework for managing multiple AI agents and handling complex conversations+2
axflowThe TypeScript framework for AI development+2
superduperSuperduper: End-to-end framework for building custom AI applications and agents.+1
melleaMellea is a library for writing generative programs.
pipelexDeclarative language for composable Al workflows. Devtool for agents and mere humans.
FrontierAgent🧩 FrontierAgent, our agent framework, open-sourced alongside it — native command-line TUI, ReAct and Agent Team modes, one command on macOS and Linux, no preinstall, no hard Docker dependency.
parlantBuild reliable customer-facing AI agents with Parlant: an interaction control harness optimized for controlled, consistent, and predictable LLM interactions.
AutoChainAutoChain: Build lightweight, extensible, and testable LLM Agents
agentaoLocal-first, governed AI agent runtime for Python — embed it in your app, or run it as a CLI or ACP server. Permissions, MCP, memory and audit replay built in.
CopilotKitThe Frontend Stack for Agents & Generative UI. React, Angular, Mobile, Slack, and more. Makers of the AG-UI Protocol+28
rocketride-serverHigh-performance AI pipeline engine with a C++ core and 50+ Python-extensible nodes. Build, debug, and scale LLM workflows with 13+ model providers, 8+ vector databases, and agent orchestration, all f+22
beeai-frameworkBuild production-ready AI agents in both Python and Typescript.+17
camel🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org+16
elizaOpen source agentic operating system+15
bishengBISHENG is an open LLM devops platform for next generation Enterprise AI applications. Powerful and comprehensive features include: GenAI workflow, RAG, Agent, Unified model management, Evaluation, SF+13
PraisonAIPraisonAI 🦞 — Hire a 24/7 AI Workforce. Stop writing boilerplate and start shipping autonomous self-improving agents that research, plan, code, and execute tasks. Deployed in 5 lines of code with bui+12
AdalFlowAdalFlow: The library to build & auto-optimize LLM applications.+9
app-controllerApp-Controller: Allow users to manipulate your App with natural language+9
LangBotProduction-grade platform for building agentic IM bots - 生产级多平台智能机器人开发平台/ Agent、知识库编排、插件系统 / Bots for Discord / Slack / LINE / Telegram / WeChat(企业微信, 企微智能机器人, 公众号) / 飞书 / 钉钉 / QQ / Matrix e.g. Integr+8
team-of-ai-agentsOpen-source framework to make AI agents' team collaboration as effective as human collaboration.+7
agent-frameworkA framework for building, orchestrating and deploying AI agents and multi-agent workflows with support for Python and .NET.+6
evolving-agentsYour toolkit for autonomous, evolving agent ecosystems. Create, execute, govern, and evolve agents that learn from experience, collaborate, build new capabilities, and operate safely within defined gu+5
BinduBindu: The identity, communication, and payments layer for AI agents.+4
MetaGPT🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming+4
ragbitsBuilding blocks for rapid development of GenAI applications +4
wavefront🔥🔥🔥 Enterprise AI middleware, alternative to unifyapps, n8n, lyzr+4
hiveMulti-Agent Harness for Production AI+3
chainlitBuild Conversational AI in minutes ⚡️+2
Hermes AgentAn open-source agent that grows with you — learns your projects, builds its own skills, and reaches you wherever you are. By Nous Research.+2
minion-agentA simple agent framework that's capable of browser use + mcp + auto instrument + plan + deep research + more+2
agentosGive agents an operating system as a library. Runs in your existing backend – no sandboxes, VMs, or SaaS. Powered by WebAssembly & V8 isolates.+1
mcp-agentBuild effective agents using Model Context Protocol and simple workflow patterns+1
ten-framework Open-source framework for conversational voice AI agents+1
dapr-agentsBuild autonomous, resilient and observable AI agents with built-in workflow orchestration, security, statefulness and telemetry.
autocontexta recursive self-improving harness designed to help your agents (and future iterations of those agents) succeed on any task
RasaGPT💬 RasaGPT is the first headless LLM chatbot platform built on top of Rasa and Langchain. Built w/ Rasa, FastAPI, Langchain, LlamaIndex, SQLModel, pgvector, ngrok, telegram
vocode-core🤖 Build voice-based LLM agents. Modular + open source.
agentpoolA unified agent orchestration hub that lets you configure and manage multiple AI agents (native, ACP, AGUI, Claude Code) via YAML, and exposes them through standardized protocols (ACP/OpenCode Server)
agentsA framework for building realtime voice AI agents 🤖🎙️📹
fast-agentCode, Build and Evaluate agents - excellent Model and Skills/MCP/ACP/A2A Support
golemGolem Cloud is the agent-native platform for building AI agents and distributed applications that never lose state, never duplicate work, and never require you to build infrastructure.
IoAAn open-source framework for collaborative AI agents, enabling diverse, distributed agents to team up and tackle complex tasks through internet-like connectivity.
langstreamBuild robust LLM applications with true composability 🔗
ROMARecursive-Open-Meta-Agent v0.1 (Beta). A meta-agent framework to build high-performance multi-agent systems.
piAI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI
waku-agentWaku Waku! Waku Agent is a local-first AI agent harness you actually own, including loop, memory, eval, all in code built to stay legible as it grows.
bolnaEnd-to-end platform for building voice first multimodal agents
owl🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation